A Comparison of Three Silver-containing Dressings in the Treatment of Infected, Chronic Wounds
Address correspondence to: Jose Verdu, PhD University of Alicante Ap. De Correos, 99 03080, Alicante Spain Phone: 0034 64 997 8317 E-mail: firstname.lastname@example.org
Abstract: Objective. To compare 3 types of silver dressing in the typical clinical conditions of a community health center, regarding the time to achieve resolution of clinical signs of local infection, and wound healing progress over 8 weeks. Methods. A prospective, comparative study involving 75 patients with infected chronic wounds who were divided into 3 treatment groups: Acticoat™ (group 1); Comfeel® Ag hydrocolloid/Biatain® Ag polyurethane foam (group 2); and Aquacel® Ag (group 3). Results. The groups were comparable at baseline. Clinical signs of infection were resolved faster in group 1 than in the other two groups (P < 0.05, median: group 1 = 2 weeks; group 2 = 4 weeks; group 3 = 4 weeks. Group 1 required fewer treatments to eliminate the clinical signs of infection (median: group 1 = 6 treatments; group 2 = 12 treatments; group 3 = 12 treatments). Patients in group 1 healed faster than patients in the other 2 groups (P < 0.05). Conclusion. The treatment in group 1 was more effective than that of groups 2 and 3 in the treatment of infected, chronic wounds. Clinical signs of infection were resolved faster (P < 0.05) and wounds healed more quickly (P < 0.05) in group 1 than in the other 2 groups. The majority of chronic wounds, typically up to 70%, can be healed within 12 to 24 weeks, but this still leaves a considerable proportion that heal with difficulty, even when advanced techniques are used.1 Although it has been shown that the presence of low levels of bacteria can actually enhance wound healing, the healing process is undoubtedly affected by the presence of high levels of bacteria—wounds can be colonized without showing any signs or symptoms of infection, and with little effect on healing, apart from a delay in healing time.2 However, clinical infection will affect wound healing and often prevents closure.3 Various antimicrobial products are available for bacterial load management, and there is significant variability in clinical practice regarding product selection. Choosing the correct antimicrobial dressing has become a significant challenge.4 It is necessary to improve the diagnosis of infection in order to use these dressings correctly.5 Silver is bactericidal against a large number of Gram-positive and Gram-negative micro-organisms, both aerobic and anaerobic, and against several multi-resistant microorganisms.6,7 Currently, there are more than 10 silver dressings on the market,8 each with different formulations (creams, foams, hydrogels, hydrocolloids, polymer films and meshes). Each formulation offers different advantages,9 some require a secondary gauze dressing or a treatment in a moist environment, while others act as a secondary dressing. Also, great differences exist with respect to clinical efficacy. It is important to understand whether the action of the dressing is that of a true antimicrobial intervention within the wound environment, or whether the level of silver is only sufficient to keep the dressing “microbiologically clean.” The activity of a silver-containing dressing is related to the amount, type, and distribution of silver in the dressing.8–10 Studies suggest that the antimicrobial activity of silver improves healing,7 but actual research findings are inconclusive or contradictory.6,7,9,11 Many of these are in-vitro studies or use animal models; clinical trials are only now beginning to appear. No other studies were found that compare different silver dressings in the treatment of chronic wounds. The objective of this study was to compare 3 types of silver dressing in typical clinical conditions encountered in a health center, focusing on time to resolution of the clinical signs of local infection, and wound healing progress over an 8-week period.
Materials and Methods
Patients. Seventy-five patients were enrolled (1 wound per patient), 25 per group. Of these, 45 (60%) were women and 30 (40%) were men, and were distributed evenly in all groups (16 women in group 1, 14 women in group 2, and 15 women in group 3). In group 2, 16 (64%) wounds were treated with Biatain Ag and 9 (36%) with Comfeel Ag. Separate analysis was not performed due to the small size of the groups. A prospective, comparative study was designed. Patients with chronic wounds showing signs of local infection were assigned to 3 treatment groups: Group 1. Acticoat™ (Smith and Nephew, Largo, Fla): treated with a nanocrystalline silver system (Acticoat consists of 3 layers: an absorbent inner core sandwiched between outer layers of silver coated, low-adherent polyethylene net), independent of the quantity of exudate present in the wounds, until the signs of local infection resolve. Polyurethane foam was used as a secondary dressing. Subsequently, treatment was continued with polyurethane foam until healing or for a maximum of 8 weeks. Group 2. Comfeel™/Biatain™ Ag (Coloplast, Minneapolis, Minn): treated with ionic silver dressings (Comfeel Ag, a hydrocolloid dressing for wounds with minimal exudate, and Biatain Ag, a polyurethane foam for wounds with moderate to high exudate. Outside of Spain, Comfeel/Biatain Ag are known as Contreet® Hydrocolloid/Contreet® Foam). Treatment was then continued with the same range of dressings without silver until healing or for a maximum of 8 weeks. Group 3. Aquacel® Ag (ConvaTec, Skillman, NJ): treated with an ionic silver dressing (Aquacel is a gelling Hydrofiber®) independent of the quantity of exudate present in the wounds, until the signs of local infection disappeared. Polyurethane foam was used as a secondary dressing. Treatment was then continued with polyurethane foam until healing, or for a maximum of 8 weeks. The dressings were changed 3 times a week. Sample size. Applying an alpha risk (a = 0.05) and beta risk (b = 0.05), 24 subjects in each group were required to detect a minimum difference of 1.3 weeks in the time to resolve infection between 2 groups, assuming 3 groups and a standard deviation of 1.05 weeks (from a pilot study with 5 wounds per group). A follow-up loss rate of 10% was assumed. Inclusion criteria. Patients presenting at the health center with a chronic wound displaying at least 3 of the following signs of local infection: pain, redness, heat, edema, and/or purulent exudate. In addition, patients had to be available for treatment over 8 weeks. Exclusion criteria. Patients who were less than 18-years-old were excluded. Patients with fever, taking antibiotics at the time of the study, those with known allergies to the dressings being used in the study, or those with acute wounds, were excluded. Patients with terminal illness were also excluded. Procedure. Eligible patients were assigned to the treatment group sequentially. The first 25 patients were assigned to group 1, the next 25 to group 2, and the final 25 patients to group 3. All patients were treated either in the home or in social healthcare centers by professionals from the Pinillo Chico Primary Care Center (El Puerto de Santa María, Cádiz, Spain), as part of their normal care duties for their assigned population. Due to the distinct appearance of each product, it was not possible to blind the professionals or the patients. Statistical analysis was blinded as to the nature of each study group. Patient follow-up was carried out until the wounds completely healed or for a maximum period of 8 weeks. Variables. Time to resolve all signs of local infection (not just a resolution of 1 or 2 signs), number of dressings required to resolve signs of local infection, Pressure Ulcer Scale for Healing (PUSH) index, time required to achieve total wound healing, occurrence of adverse effects in the wound bed or periwound skin related to the dressings (maceration, erythema, excoriation, vesicles, pain when changing the dressing, edema, staining of the wound bed or periwound skin, and the remains of the dressing in the wound after removal). Statistical analysis. Central trend measurements were used for the descriptive analysis (mean, median, standard deviation, and minimum and maximum values) of the quantitative variables, and for frequencies and proportions of qualitative variables. The level of significance for the comparative analysis was 0.05, which was adjusted using the Benjamini and Hochberg procedure, where 2 comparisons were made between treatment groups (group 1 versus groups 2 and 3). Kaplan-Meier plots were used to compare the probability of resolving infection and the probability of healing during the course of the study between treatment groups. Multivariate Cox regression models were used to test for a difference in the time to resolution of infection and the healing time between group 1 and groups 2 and 3, after adjusting for baseline factors that were found to be significantly associated with these outcomes. The SPSS package (SPSS Inc., Chicago, Ill) was used in the statistical analysis.
Baseline characteristics. The majority of the wounds were leg ulcers (n = 50) followed by pressure ulcers (n = 15), diabetic foot ulcers (n = 7), and post-traumatic ulcers (n = 3). The distribution by wound type and treatment group is shown in Table 1. Baseline characteristics of the patients and wounds are shown in Table 2. Groups such as this, with fewer than 30 cases and no normal distribution, are better represented by the median, rather than the mean. According to median values, the groups were comparable; the mean values were more sensitive to extreme values in the population. All wounds showed redness and heat. Sixty patients (80%) experienced pain, (group 1 = 20, group 2 = 21, and group 3 = 19). Thirty-six wounds (48%) had purulent exudate (group 1 = 15, group 2 = 6, and group 3 = 15). Therefore, statistically significant differences at baseline regarding this variable favor group 2 (P = 0.013). Resolution of clinical signs of infection. All clinical signs of infection resolved significantly faster in group 1 compared to the other two groups (P < 0.05, Cox regression proportional hazard). The clinical signs of infection were nearly twice as likely to resolve in group 1 than in the other 2 groups at any time during the first 4 weeks of treatment: hazard ratio, group 1:group 2 = 1.96 (P = 0.018; CI: 95% = 1.12–3.43); and hazard ratio, group 1:group 3 = 1.89 (P = 0.025; CI: 95% = 1.08–3.31). There was no evidence that any other baseline factors were associated with the time to resolution of infection. The time to resolution of infection and the number of dressing applications required are summarized in Table 3. The time to resolution of infection was less in group 1 (median: group 1 = 2 weeks, group 2 = 4 weeks, group 3 = 4 weeks), and fewer treatments were required to eliminate clinical signs of infection in group 1 (median: group 1 = 6 treatments; group 2 = 12 treatments; group 3 = 12 treatments). The number of wounds in each of the first 4 weeks that no longer showed clinical signs of local infection is shown in Table 4. The probability of resolving the clinical signs of infection during the course of the study are illustrated in Kaplan-Meier plots (Figure 1). Complete healing. Time to complete healing was significantly faster in group 1 compared to the other groups (P < 0.05), after adjusting for baseline wound area. Patients in group 1 were nearly 3 times more likely to heal at any time during the study than patients in groups 2 or 3. Hazard ratios. Group 1:group 2 = 2.74 (P = 0.042; CI: 95% = 1.04–7.27); group 1:group 3 = 2.80 (P = 0.027; CI: 95% = 1.13–6.94). Time to healing increased significantly with baseline wound area (P = 0.029, hazard ratio = 0.90; P = 0.029; CI: 95% = 0.82–0.99). The percentage of wounds healed during the study is shown in Figure 2. The number of wounds healed from weeks 5–8 is shown in Table 5. The differences between treatment groups in the probability of healing during the course of the study are illustrated in the Kaplan-Meier plots (Figure 3). Unwanted effects. During the study, 189 dressing changes were carried out in group 1, 291 in group 2, and 285 in group 3. During dressing changes, unwanted effects were recorded (Table 6). In all dressings, neither pain during dressing changes nor window edema was observed.
The results of this study confirm that silver dressings are effective in eliminating local infection, and therefore, encourage healing. However, there are differences in outcomes. The authors could not find any published studies that compare different silver dressings in the treatment of chronic wounds in the clinical practice setting. The systematic review by Vermeulen et al11 only found 3 clinical trials comparing products, and only the CONTOP12 study compares, indirectly, other silver dressings, but of unspecified composition. The results of the present study cannot be compared to those of the CONTOP study, since they are methodologically different. There are some limitations to this study. While the group size may have been small, the sample size for the primary variable (resolution of the clinical signs of infection) was adequate. The study is not statistically powered to detect differences in secondary variables, but where statistically significant differences were detected, they were valid. The authors debated whether to obtain bacteriological samples, but according to the EWMA documents on wound infection,5,13 bacteriological analysis alone does not establish the presence of infection, and subtle changes in clinical signs of infection are preferred. Ideally, follow up would have been for at least 6 months, but this is costly and there is usually a high rate of patient attrition over such a long period. The clinical practices applied in this study corresponded closely to EWMA recommendations13 to use a topical antimicrobial for infected stage 2 or 3 wounds, and when the obvious signs of infection are eliminated, to discontinue the antimicrobial treatment and continue treatment based on local protocol. There is debate as to how long an antimicrobial treatment should be used. In the present study, remission of clinical signs of local infection occurred between 2 and 4 weeks. Other studies14–16 show that a 20%–40% reduction of the wound surface in the first 2 to 4 weeks of treatment is predictive of healing. However, the less time an antimicrobial treatment is used the less chance there is of inducing microbial resistance, thus one should choose the treatment that achieves results in the shortest time. Leaper9 concluded that nanocrystalline silver had the highest level of evidence and superior outcomes. Sibbald et al17 reported that nanocrystalline silver significantly reduced total bacterial counts in chronic venous ulcers, a finding that has not been seen with any other product. Recent work18 has shown that true cost savings can be achieved by targeting wounds that need antimicrobial intervention and using the product for an appropriate length of time, even though the initial dressing price may be higher. The results of the present study could also support this conclusion.
In this study, Acticoat was found to be more effective than Comfeel/Biatain and Aquacel Ag in terms of time to resolve clinical signs of local infection and time to wound healing. Clinical signs of infection were nearly twice as likely to resolve during the first 4 weeks of treatment with Acticoat. Patients in the Acticoat group were nearly 3 times more likely to heal at any time during the study than patients in the Comfeel/Biatain and Aquacel Ag groups.
From the CS Pinillo Chico, Cádiz, Spain; Hospital Puerto Real, Cádiz, Spain; CS Sanlucar de Barrameda, Cádiz, Spain; University of Alicante, Spain; Interdisciplinary Chronic Wounds Unit, Departamento de Salud 20, Alicante, Spain
Disclosure: This study was supported by an unrestricted research grant from Smith & Nephew, Spain.